© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 19585
    Prev
    Next
    INDEX
    Explanations

    information related to educational programs and traffic regulations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    756
    +0.15
    0.7%
    570
    +0.10
    0.5%
    51
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.15
    -0.02
    570
    +0.10
    0.01
    680
    +0.10
    0.03
    Negative Logits
     whilst
    -0.90
     Whilst
    -0.83
     jew
    -0.77
     imperialist
    -0.77
     traitor
    -0.75
    Picture
    -0.75
     evil
    -0.73
     Illuminati
    -0.72
    ↵Âł
    -0.72
    alien
    -0.72
    POSITIVE LOGITS
     backlog
    1.23
     staffing
    1.13
     audits
    1.11
     statewide
    1.04
     rollout
    1.02
     phased
    0.97
     manpower
    0.96
     inspections
    0.94
     enrollment
    0.93
     supervisors
    0.93
    Activations Density 0.945%

    No Known Activations